| Code | i3WaterSDC11 |
|---|---|
| Host Institution | Universitat Politècnica de Catalunya Research Center for Supervision, Safety and Automatic Control (Universitat Politècnica de Catalunya) |
| Location | Terrassa, 08222 |
| Supervisor(s) | Main Supervisor: Dr Ramon Pérez Magrané (UPC, Spain) Co-supervisor: Dr Mario Castro-Gama (VIT, The Netherlands) Industrial Mentor – Experts: Mr Sergi Grau Torrent (AM, Spain) |
| Research Field | Artificial Intelligence |
| Contract type | Fixed term contract — Full time, 37.5 h/week |
| Application Deadline | september 15th, 2026 |
Description
We are seeking a highly motivated and talented Doctoral Candidate (DC) to join our team for a PhD project titled: Distributed and centralized data mining for WDS in the context of IoT and Big data.
Modern Water Distribution Systems (WDS) are rapidly evolving into smart, data-rich environments. This research project addresses the critical challenge of processing, validating, and extracting value from heterogeneous, high-frequency data streams. The project is structured around two core technological pillars: developing decentralized Artificial Intelligence (AI) models at the edge/sensor level, and implementing centralized advanced analytics in the cloud. The ultimate goal is to fuse both approaches, creating a synergistic framework where distributed and centralized intelligence enhance each other to optimize water network management.
Key Responsibilities and Research Tasks:
The research will be divided into three main operational phases:
1.AI-based Distributed Data Validation & Reconstruction: Develop a novel methodology leveraging AI/Machine Learning models for decentralized, automated data validation and data reconstruction directly at the source (edge/sensor level) to handle missing data or sensor anomalies.
2. Heterogeneous IoT Data Analytics: Explore, design, and implement advanced analysis systems (including AI frameworks, hydraulic and quality models) tailored for SensorThings (IoT) data. You will integrate and analyze diverse data streams originating from telecontrol (SCADA), laboratory analytics, maintenance logs, and external sources.
3. Framework Integration & Synergy Analysis: Integrate both distributed and centralized approaches into a unified platform. You will conduct comparative analyses to evaluate performance trade-offs and investigate how distributed and centralized results can mutually enhance system-wide accuracy and resilience.
Expected Outcomes & Deliverables:
By the end of the PhD, the researcher is expected to have successfully developed and delivered:
- Outcome 1: A robust methodology and software framework for Distributed WDS data collection and validation at the edge.
- Outcome 2: An innovative Integration tool capable of seamlessly connecting live IoT data streams with hydraulic simulation models (e.g., EPANET).
- Outcome 3: A fully functional Hybrid architecture that features decentralized data storage (at the edge/sensor level) coupled with centralized model management (in the cloud).
Requirements
Education level
Master Degree
Skills / Qualifications
- Education: An outstanding Master’s degree (or equivalent) in Industrial Engineering, Computer Science, Data Science, AI/Machine Learning, Telecommunications Engineering, Applied Mathematics/Physics, or a related computational field.
- Technical Skills:
- Strong programming skills in Python or R.
- Solid foundation in Machine Learning and Deep Learning architectures.
- Familiarity with IoT protocols, data integration pipelines, and handling heterogeneous datasets.
- Familiarity with water networks modelling.
- Knowledge of distributed computing frameworks and decentralized databases is a strong plus.
Required languages
English – C1 / Spanish or catalan will be taken into consideration
